pori-python 0.1.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- pori_python/__init__.py +1 -0
- pori_python/graphkb/__init__.py +2 -0
- pori_python/graphkb/constants.py +220 -0
- pori_python/graphkb/genes.py +540 -0
- pori_python/graphkb/match.py +710 -0
- pori_python/graphkb/py.typed +0 -0
- pori_python/graphkb/statement.py +48 -0
- pori_python/graphkb/util.py +523 -0
- pori_python/graphkb/vocab.py +198 -0
- pori_python/ipr/__init__.py +0 -0
- pori_python/ipr/annotate.py +373 -0
- pori_python/ipr/connection.py +214 -0
- pori_python/ipr/constants.py +9 -0
- pori_python/ipr/inputs.py +593 -0
- pori_python/ipr/ipr.py +390 -0
- pori_python/ipr/main.py +512 -0
- pori_python/ipr/py.typed +0 -0
- pori_python/ipr/summary.py +412 -0
- pori_python/ipr/therapeutic_options.py +83 -0
- pori_python/ipr/util.py +173 -0
- pori_python/types.py +177 -0
- pori_python-0.1.0.dist-info/LICENSE +674 -0
- pori_python-0.1.0.dist-info/METADATA +151 -0
- pori_python-0.1.0.dist-info/RECORD +49 -0
- pori_python-0.1.0.dist-info/WHEEL +5 -0
- pori_python-0.1.0.dist-info/dependency_links.txt +1 -0
- pori_python-0.1.0.dist-info/entry_points.txt +2 -0
- pori_python-0.1.0.dist-info/top_level.txt +2 -0
- tests/__init__.py +0 -0
- tests/test_graphkb/__init__.py +0 -0
- tests/test_graphkb/data.py +78 -0
- tests/test_graphkb/test_genes.py +306 -0
- tests/test_graphkb/test_graphkb.py +32 -0
- tests/test_graphkb/test_match.py +662 -0
- tests/test_graphkb/test_statement.py +100 -0
- tests/test_graphkb/test_util.py +151 -0
- tests/test_graphkb/test_vocab.py +85 -0
- tests/test_ipr/__init__.py +0 -0
- tests/test_ipr/constants.py +3 -0
- tests/test_ipr/test_annotate.py +149 -0
- tests/test_ipr/test_connection.py +97 -0
- tests/test_ipr/test_inputs.py +295 -0
- tests/test_ipr/test_ipr.py +338 -0
- tests/test_ipr/test_main.py +154 -0
- tests/test_ipr/test_probe.py +72 -0
- tests/test_ipr/test_summary.py +157 -0
- tests/test_ipr/test_upload.py +274 -0
- tests/test_ipr/test_util.py +29 -0
- tests/test_ipr/util.py +9 -0
|
@@ -0,0 +1,295 @@
|
|
|
1
|
+
import json
|
|
2
|
+
import numpy as np
|
|
3
|
+
import os
|
|
4
|
+
import pandas as pd
|
|
5
|
+
import pytest
|
|
6
|
+
from unittest import mock
|
|
7
|
+
|
|
8
|
+
from pori_python.graphkb.match import INPUT_COPY_CATEGORIES
|
|
9
|
+
from pori_python.ipr.inputs import (
|
|
10
|
+
COPY_OPTIONAL,
|
|
11
|
+
check_comparators,
|
|
12
|
+
check_variant_links,
|
|
13
|
+
create_graphkb_sv_notation,
|
|
14
|
+
preprocess_copy_variants,
|
|
15
|
+
preprocess_expression_variants,
|
|
16
|
+
preprocess_small_mutations,
|
|
17
|
+
preprocess_structural_variants,
|
|
18
|
+
validate_report_content,
|
|
19
|
+
)
|
|
20
|
+
from pori_python.ipr.util import logger
|
|
21
|
+
from pori_python.types import IprFusionVariant, IprGeneVariant
|
|
22
|
+
|
|
23
|
+
DATA_DIR = os.path.join(os.path.dirname(__file__), "test_data")
|
|
24
|
+
NON_EMPTY_STRING_NULLS = ["", None, np.nan, pd.NA]
|
|
25
|
+
|
|
26
|
+
|
|
27
|
+
def read_data_file(filename):
|
|
28
|
+
pass
|
|
29
|
+
|
|
30
|
+
|
|
31
|
+
class TestPreProcessSmallMutations:
|
|
32
|
+
def test_load_test_file(self) -> None:
|
|
33
|
+
records = preprocess_small_mutations(
|
|
34
|
+
pd.read_csv(os.path.join(DATA_DIR, "small_mutations.tab"), sep="\t").to_dict("records")
|
|
35
|
+
)
|
|
36
|
+
assert records
|
|
37
|
+
assert len(records) == 2614
|
|
38
|
+
|
|
39
|
+
def test_maintains_optional_fields(self):
|
|
40
|
+
original = {
|
|
41
|
+
"gene": "A1BG",
|
|
42
|
+
"proteinChange": "p.V460M",
|
|
43
|
+
"zygosity": "het",
|
|
44
|
+
"tumourAltCount": 42,
|
|
45
|
+
"tumourRefCount": 48,
|
|
46
|
+
"hgvsProtein": "",
|
|
47
|
+
"transcript": "ENST1000",
|
|
48
|
+
"hgvsCds": "",
|
|
49
|
+
"hgvsGenomic": "",
|
|
50
|
+
"key": "02fe85a3477784b5ac0f8ecffb300d10",
|
|
51
|
+
"variant": "blargh",
|
|
52
|
+
"chromosome": "2",
|
|
53
|
+
"startPosition": 1234,
|
|
54
|
+
}
|
|
55
|
+
records = preprocess_small_mutations([original])
|
|
56
|
+
record = records[0]
|
|
57
|
+
assert record["variantType"] == "mut"
|
|
58
|
+
for col in original:
|
|
59
|
+
assert col in record
|
|
60
|
+
assert record["variant"] == "A1BG:p.V460M"
|
|
61
|
+
assert "endPosition" in record
|
|
62
|
+
assert record["endPosition"] == record["startPosition"]
|
|
63
|
+
assert "tumourDepth" in record
|
|
64
|
+
assert record["tumourDepth"] == 90
|
|
65
|
+
|
|
66
|
+
def test_null(self):
|
|
67
|
+
original = {
|
|
68
|
+
"gene": "A1BG",
|
|
69
|
+
"proteinChange": "p.V460M",
|
|
70
|
+
"tumourAltCount": 42,
|
|
71
|
+
"tumourRefCount": 48,
|
|
72
|
+
"startPosition": 1234,
|
|
73
|
+
}
|
|
74
|
+
# Make sure TEST_KEYS are appropriate.
|
|
75
|
+
# For some fields, like 'ref' and 'alt', NA is _not_ equivalent to a null string.
|
|
76
|
+
TEST_KEYS = ["startPosition", "endPosition", "tumourAltCount", "tumourRefCount"]
|
|
77
|
+
for key in TEST_KEYS:
|
|
78
|
+
for null in NON_EMPTY_STRING_NULLS:
|
|
79
|
+
small_mut = original.copy()
|
|
80
|
+
small_mut[key] = null
|
|
81
|
+
records = preprocess_small_mutations([small_mut])
|
|
82
|
+
record = records[0]
|
|
83
|
+
assert record["variantType"] == "mut"
|
|
84
|
+
for col in original:
|
|
85
|
+
assert col in record
|
|
86
|
+
assert record["variant"] == "A1BG:p.V460M"
|
|
87
|
+
assert "endPosition" in record
|
|
88
|
+
|
|
89
|
+
def test_load_small_mutations_probe(self) -> None:
|
|
90
|
+
records = preprocess_small_mutations(
|
|
91
|
+
pd.read_csv(os.path.join(DATA_DIR, "small_mutations_probe.tab"), sep="\t").to_dict(
|
|
92
|
+
"records"
|
|
93
|
+
)
|
|
94
|
+
)
|
|
95
|
+
assert records
|
|
96
|
+
assert len(records) == 4
|
|
97
|
+
assert records[0]["variantType"] == "mut"
|
|
98
|
+
assert "variant" in records[0]
|
|
99
|
+
|
|
100
|
+
|
|
101
|
+
class TestPreProcessCopyVariants:
|
|
102
|
+
def test_load_copy_variants(self) -> None:
|
|
103
|
+
records = preprocess_copy_variants(
|
|
104
|
+
pd.read_csv(os.path.join(DATA_DIR, "copy_variants.tab"), sep="\t").to_dict("records")
|
|
105
|
+
)
|
|
106
|
+
assert records
|
|
107
|
+
assert len(records) == 4603
|
|
108
|
+
assert records[0]["variantType"] == "cnv"
|
|
109
|
+
assert "variant" in records[0]
|
|
110
|
+
|
|
111
|
+
def test_null(self):
|
|
112
|
+
for kb_cat in list(INPUT_COPY_CATEGORIES.values()) + NON_EMPTY_STRING_NULLS:
|
|
113
|
+
original = {"gene": "ERBB2", "kbCategory": kb_cat}
|
|
114
|
+
for key in COPY_OPTIONAL:
|
|
115
|
+
for null in NON_EMPTY_STRING_NULLS:
|
|
116
|
+
copy_var = original.copy()
|
|
117
|
+
copy_var[key] = null
|
|
118
|
+
records = preprocess_copy_variants([copy_var])
|
|
119
|
+
record = records[0]
|
|
120
|
+
assert record["variantType"] == "cnv"
|
|
121
|
+
|
|
122
|
+
|
|
123
|
+
def test_load_structural_variants() -> None:
|
|
124
|
+
records = preprocess_structural_variants(
|
|
125
|
+
pd.read_csv(os.path.join(DATA_DIR, "fusions.tab"), sep="\t").to_dict("records")
|
|
126
|
+
)
|
|
127
|
+
assert records
|
|
128
|
+
assert len(records) == 7
|
|
129
|
+
assert records[0]["variantType"] == "sv"
|
|
130
|
+
assert "variant" in records[0]
|
|
131
|
+
|
|
132
|
+
|
|
133
|
+
def test_load_expression_variants() -> None:
|
|
134
|
+
records = preprocess_expression_variants(
|
|
135
|
+
pd.read_csv(os.path.join(DATA_DIR, "expression.tab"), sep="\t").to_dict("records")
|
|
136
|
+
)
|
|
137
|
+
assert records
|
|
138
|
+
assert len(records) == 4603
|
|
139
|
+
assert records[0]["variantType"] == "exp"
|
|
140
|
+
assert "variant" in records[0]
|
|
141
|
+
|
|
142
|
+
|
|
143
|
+
class TestCheckVariantLinks:
|
|
144
|
+
def test_sm_missing_copy_empty_ok(self) -> None:
|
|
145
|
+
genes = check_variant_links(
|
|
146
|
+
small_mutations=[IprGeneVariant({"gene": "KRAS"})], # type: ignore
|
|
147
|
+
copy_variants=[],
|
|
148
|
+
expression_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
149
|
+
structural_variants=[],
|
|
150
|
+
)
|
|
151
|
+
assert genes == {"KRAS"}
|
|
152
|
+
|
|
153
|
+
def test_sm_missing_exp_empty_ok(self) -> None:
|
|
154
|
+
genes = check_variant_links(
|
|
155
|
+
small_mutations=[IprGeneVariant({"gene": "KRAS"})], # type: ignore
|
|
156
|
+
copy_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
157
|
+
expression_variants=[],
|
|
158
|
+
structural_variants=[],
|
|
159
|
+
)
|
|
160
|
+
assert genes == {"KRAS"}
|
|
161
|
+
|
|
162
|
+
def test_sm_missing_copy(self) -> None:
|
|
163
|
+
with mock.patch.object(logger, "debug") as mock_debug:
|
|
164
|
+
check_variant_links(
|
|
165
|
+
small_mutations=[IprGeneVariant({"gene": "KRAS"})], # type: ignore
|
|
166
|
+
copy_variants=[IprGeneVariant({"gene": "CDK", "variant": ""})], # type: ignore
|
|
167
|
+
expression_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
168
|
+
structural_variants=[],
|
|
169
|
+
)
|
|
170
|
+
assert mock_debug.called
|
|
171
|
+
|
|
172
|
+
def test_sm_missing_exp(self) -> None:
|
|
173
|
+
with mock.patch.object(logger, "debug") as mock_debug:
|
|
174
|
+
check_variant_links(
|
|
175
|
+
small_mutations=[IprGeneVariant({"gene": "KRAS"})], # type: ignore
|
|
176
|
+
copy_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
177
|
+
expression_variants=[IprGeneVariant({"gene": "CDK", "variant": ""})], # type: ignore
|
|
178
|
+
structural_variants=[],
|
|
179
|
+
)
|
|
180
|
+
assert mock_debug.called
|
|
181
|
+
|
|
182
|
+
def test_with_valid_inputs(self) -> None:
|
|
183
|
+
genes = check_variant_links(
|
|
184
|
+
small_mutations=[IprGeneVariant({"gene": "KRAS"})], # type: ignore
|
|
185
|
+
copy_variants=[
|
|
186
|
+
IprGeneVariant({"gene": "KRAS", "variant": ""}), # type: ignore
|
|
187
|
+
IprGeneVariant({"gene": "CDK", "variant": ""}), # type: ignore
|
|
188
|
+
],
|
|
189
|
+
expression_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
190
|
+
structural_variants=[],
|
|
191
|
+
)
|
|
192
|
+
assert genes == {"KRAS"}
|
|
193
|
+
|
|
194
|
+
def test_copy_missing_exp(self) -> None:
|
|
195
|
+
with mock.patch.object(logger, "debug") as mock_debug:
|
|
196
|
+
check_variant_links(
|
|
197
|
+
small_mutations=[],
|
|
198
|
+
copy_variants=[
|
|
199
|
+
IprGeneVariant({"gene": "BRAF", "variant": "copy gain"}), # type: ignore
|
|
200
|
+
IprGeneVariant({"gene": "KRAS", "variant": ""}), # type: ignore
|
|
201
|
+
],
|
|
202
|
+
expression_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
203
|
+
structural_variants=[],
|
|
204
|
+
)
|
|
205
|
+
assert mock_debug.called
|
|
206
|
+
|
|
207
|
+
def test_exp_missing_copy(self) -> None:
|
|
208
|
+
with mock.patch.object(logger, "debug") as mock_debug:
|
|
209
|
+
check_variant_links(
|
|
210
|
+
small_mutations=[],
|
|
211
|
+
copy_variants=[IprGeneVariant({"gene": "KRAS", "variant": ""})], # type: ignore
|
|
212
|
+
expression_variants=[
|
|
213
|
+
IprGeneVariant({"gene": "BRAF", "variant": "increased expression"}) # type: ignore
|
|
214
|
+
],
|
|
215
|
+
structural_variants=[],
|
|
216
|
+
)
|
|
217
|
+
assert mock_debug.called
|
|
218
|
+
|
|
219
|
+
|
|
220
|
+
class TestCreateGraphkbSvNotation:
|
|
221
|
+
def test_both_genes_and_exons(self) -> None:
|
|
222
|
+
notation = create_graphkb_sv_notation(
|
|
223
|
+
IprFusionVariant({"gene1": "A", "gene2": "B", "exon1": 1, "exon2": 2}) # type: ignore
|
|
224
|
+
)
|
|
225
|
+
assert notation == "(A,B):fusion(e.1,e.2)"
|
|
226
|
+
|
|
227
|
+
def test_one_exon_missing(self) -> None:
|
|
228
|
+
notation = create_graphkb_sv_notation(
|
|
229
|
+
IprFusionVariant({"gene1": "A", "gene2": "B", "exon1": "", "exon2": 2}) # type: ignore
|
|
230
|
+
)
|
|
231
|
+
assert notation == "(A,B):fusion(e.?,e.2)"
|
|
232
|
+
|
|
233
|
+
def test_one_gene_missing(self) -> None:
|
|
234
|
+
notation = create_graphkb_sv_notation(
|
|
235
|
+
IprFusionVariant({"gene1": "A", "gene2": "", "exon1": 1, "exon2": 2}) # type: ignore
|
|
236
|
+
)
|
|
237
|
+
assert notation == "(A,?):fusion(e.1,e.2)"
|
|
238
|
+
|
|
239
|
+
def test_first_gene_missing(self) -> None:
|
|
240
|
+
notation = create_graphkb_sv_notation(
|
|
241
|
+
IprFusionVariant({"gene1": "", "gene2": "B", "exon1": 1, "exon2": 2}) # type: ignore
|
|
242
|
+
)
|
|
243
|
+
assert notation == "(B,?):fusion(e.2,e.1)"
|
|
244
|
+
|
|
245
|
+
def test_no_genes_error(self) -> None:
|
|
246
|
+
with pytest.raises(ValueError):
|
|
247
|
+
create_graphkb_sv_notation(
|
|
248
|
+
IprFusionVariant({"gene1": "", "gene2": "", "exon1": 1, "exon2": 2, "key": "x"}) # type: ignore
|
|
249
|
+
)
|
|
250
|
+
|
|
251
|
+
|
|
252
|
+
class TestCheckComparators:
|
|
253
|
+
def test_missing_disease_expression_error(self):
|
|
254
|
+
content = {"comparators": [{"analysisRole": "expression (primary site)"}]}
|
|
255
|
+
variants = [{}]
|
|
256
|
+
|
|
257
|
+
with pytest.raises(ValueError):
|
|
258
|
+
check_comparators(content, variants)
|
|
259
|
+
|
|
260
|
+
def test_missing_primary_expression_error(self):
|
|
261
|
+
content = {"comparators": [{"analysisRole": "expression (disease)"}]}
|
|
262
|
+
variants = [{"primarySiteFoldChange": 1}]
|
|
263
|
+
|
|
264
|
+
with pytest.raises(ValueError):
|
|
265
|
+
check_comparators(content, variants)
|
|
266
|
+
|
|
267
|
+
def test_missing_biopsy_expression_error(self):
|
|
268
|
+
content = {"comparators": [{"analysisRole": "expression (disease)"}]}
|
|
269
|
+
variants = [{"biopsySitePercentile": 1}]
|
|
270
|
+
|
|
271
|
+
with pytest.raises(ValueError):
|
|
272
|
+
check_comparators(content, variants)
|
|
273
|
+
|
|
274
|
+
def test_expression_not_required_without_variants(self):
|
|
275
|
+
content = {"comparators": []}
|
|
276
|
+
variants = []
|
|
277
|
+
|
|
278
|
+
assert check_comparators(content, variants) is None
|
|
279
|
+
|
|
280
|
+
def test_missing_mutation_burden(self):
|
|
281
|
+
content = {
|
|
282
|
+
"comparators": [{"analysisRole": "mutation burden (secondary)"}],
|
|
283
|
+
"images": [{"key": "mutationBurden.density_snv.primary"}],
|
|
284
|
+
}
|
|
285
|
+
variants = []
|
|
286
|
+
|
|
287
|
+
with pytest.raises(ValueError):
|
|
288
|
+
check_comparators(content, variants)
|
|
289
|
+
|
|
290
|
+
|
|
291
|
+
@pytest.mark.parametrize("example_name", ["no_variants", "sm_and_exp", "sm_only"])
|
|
292
|
+
def test_valid_json_inputs(example_name: str):
|
|
293
|
+
with open(os.path.join(DATA_DIR, "json_examples", f"{example_name}.json"), "r") as fh:
|
|
294
|
+
content = json.load(fh)
|
|
295
|
+
validate_report_content(content)
|
|
@@ -0,0 +1,338 @@
|
|
|
1
|
+
import pytest
|
|
2
|
+
from unittest.mock import Mock, patch
|
|
3
|
+
|
|
4
|
+
from pori_python.graphkb import statement as gkb_statement
|
|
5
|
+
from pori_python.graphkb import vocab as gkb_vocab
|
|
6
|
+
from pori_python.ipr.ipr import convert_statements_to_alterations, germline_kb_matches
|
|
7
|
+
from pori_python.types import Statement
|
|
8
|
+
|
|
9
|
+
DISEASE_RIDS = ["#138:12", "#138:13"]
|
|
10
|
+
APPROVED_EVIDENCE_RIDS = ["approved1", "approved2"]
|
|
11
|
+
GERMLINE_VARIANTS = [
|
|
12
|
+
{
|
|
13
|
+
"key": "1",
|
|
14
|
+
"germline": True,
|
|
15
|
+
"hgvsCds": "SLC28A3:c.1381C>T",
|
|
16
|
+
"hgvsGenomic": "chr9:g.84286011G>A",
|
|
17
|
+
"hgvsProtein": "SLC28A3:p.L461L",
|
|
18
|
+
"ncbiBuild": "GRCh38",
|
|
19
|
+
"normalAltCount": 37,
|
|
20
|
+
"normalDepth": 37,
|
|
21
|
+
"normalRefCount": 0,
|
|
22
|
+
"proteinChange": "p.L461L",
|
|
23
|
+
"rnaAltCount": "",
|
|
24
|
+
"rnaDepth": "",
|
|
25
|
+
"rnaRefCount": "",
|
|
26
|
+
"startPosition": 84286011,
|
|
27
|
+
"transcript": "ENST00000376238",
|
|
28
|
+
"tumourAltCount": "",
|
|
29
|
+
"tumourDepth": "",
|
|
30
|
+
"tumourRefCount": "",
|
|
31
|
+
"variant": "SLC28A3:p.L461L",
|
|
32
|
+
"variantType": "mut",
|
|
33
|
+
"zygosity": "",
|
|
34
|
+
},
|
|
35
|
+
{
|
|
36
|
+
"key": "2",
|
|
37
|
+
"germline": True,
|
|
38
|
+
"hgvsCds": "BRCA1:c.4837A>",
|
|
39
|
+
"hgvsGenomic": "chr17:g.43071077T>C",
|
|
40
|
+
"hgvsProtein": "BRCA1:p.S1613G",
|
|
41
|
+
"normalAltCount": 33,
|
|
42
|
+
"normalDepth": 33,
|
|
43
|
+
"normalRefCount": 0,
|
|
44
|
+
"tumourAltCount": 37,
|
|
45
|
+
"tumourDepth": 37,
|
|
46
|
+
"tumourRefCount": 0,
|
|
47
|
+
},
|
|
48
|
+
]
|
|
49
|
+
|
|
50
|
+
SOMATIC_VARIANTS = [
|
|
51
|
+
{
|
|
52
|
+
"key": "1",
|
|
53
|
+
"gene": "SLC28A3",
|
|
54
|
+
"germline": False,
|
|
55
|
+
"hgvsCds": "SLC28A3:c.1381C>T",
|
|
56
|
+
"hgvsGenomic": "chr9:g.84286011G>A",
|
|
57
|
+
"hgvsProtein": "SLC28A3:p.L461L",
|
|
58
|
+
"ncbiBuild": "GRCh38",
|
|
59
|
+
"normalAltCount": 0,
|
|
60
|
+
"normalDepth": 37,
|
|
61
|
+
"normalRefCount": 37,
|
|
62
|
+
"tumourAltCount": 37,
|
|
63
|
+
"tumourDepth": 37,
|
|
64
|
+
"tumourRefCount": 0,
|
|
65
|
+
"variant": "SLC28A3:p.L461L",
|
|
66
|
+
"variantType": "mut",
|
|
67
|
+
"zygosity": "",
|
|
68
|
+
},
|
|
69
|
+
{
|
|
70
|
+
"key": "2",
|
|
71
|
+
"germline": False,
|
|
72
|
+
"hgvsCds": "BRCA1:c.4837A>",
|
|
73
|
+
"hgvsGenomic": "chr17:g.43071077T>C",
|
|
74
|
+
"hgvsProtein": "BRCA1:p.S1613G",
|
|
75
|
+
"normalAltCount": 1,
|
|
76
|
+
"normalDepth": 33,
|
|
77
|
+
"normalRefCount": 32,
|
|
78
|
+
"tumourAltCount": 37,
|
|
79
|
+
"tumourDepth": 37,
|
|
80
|
+
"tumourRefCount": 0,
|
|
81
|
+
},
|
|
82
|
+
]
|
|
83
|
+
|
|
84
|
+
GERMLINE_KB_MATCHES = [
|
|
85
|
+
{
|
|
86
|
+
"variant": "1",
|
|
87
|
+
"approvedTherapy": False,
|
|
88
|
+
"category": "pharmacogenomic",
|
|
89
|
+
"context": "anthracyclines",
|
|
90
|
+
"kbContextId": "#122:20944",
|
|
91
|
+
"kbRelevanceId": "#147:38",
|
|
92
|
+
"kbStatementId": "#154:13387",
|
|
93
|
+
"kbVariant": "SLC28A3:c.1381C>T",
|
|
94
|
+
"kbVariantId": "#159:5426",
|
|
95
|
+
"matchedCancer": False,
|
|
96
|
+
"reference": "PMID: 27197003",
|
|
97
|
+
"relevance": "decreased toxicity",
|
|
98
|
+
"reviewStatus": "initial",
|
|
99
|
+
},
|
|
100
|
+
{
|
|
101
|
+
"variant": "2",
|
|
102
|
+
"approvedTherapy": True,
|
|
103
|
+
"category": "cancer predisposition",
|
|
104
|
+
"kbContextId": "#135:8764",
|
|
105
|
+
"kbRelevanceId": "#147:32",
|
|
106
|
+
"kbStatementId": "#155:13511",
|
|
107
|
+
"kbVariant": "BRCA1 mutation",
|
|
108
|
+
"kbVariantId": "#161:938",
|
|
109
|
+
"matchedCancer": False,
|
|
110
|
+
"reference": "MOAlmanac FDA-56",
|
|
111
|
+
"relevance": "therapy",
|
|
112
|
+
"reviewStatus": None,
|
|
113
|
+
},
|
|
114
|
+
]
|
|
115
|
+
|
|
116
|
+
SOMATIC_KB_MATCHES = [
|
|
117
|
+
{
|
|
118
|
+
"variant": "1",
|
|
119
|
+
"approvedTherapy": False,
|
|
120
|
+
"category": "prognostic",
|
|
121
|
+
"kbContextId": "somatic_test",
|
|
122
|
+
"kbRelevanceId": "#147:38",
|
|
123
|
+
"kbStatementId": "#154:13387",
|
|
124
|
+
"kbVariant": "SLC28A3:c.1381C>T",
|
|
125
|
+
"kbVariantId": "#159:5426",
|
|
126
|
+
"relevance": "prognostic",
|
|
127
|
+
"reviewStatus": "initial",
|
|
128
|
+
},
|
|
129
|
+
{
|
|
130
|
+
"variant": "2",
|
|
131
|
+
"approvedTherapy": True,
|
|
132
|
+
"category": "therapy",
|
|
133
|
+
"kbContextId": "#135:8764",
|
|
134
|
+
"kbRelevanceId": "#147:32",
|
|
135
|
+
"kbStatementId": "#155:13511",
|
|
136
|
+
"kbVariant": "BRCA1 mutation",
|
|
137
|
+
"kbVariantId": "#161:938",
|
|
138
|
+
"matchedCancer": False,
|
|
139
|
+
"reference": "MOAlmanac FDA-56",
|
|
140
|
+
"relevance": "therapy",
|
|
141
|
+
"reviewStatus": None,
|
|
142
|
+
},
|
|
143
|
+
]
|
|
144
|
+
|
|
145
|
+
|
|
146
|
+
@pytest.fixture
|
|
147
|
+
def graphkb_conn():
|
|
148
|
+
class QueryMock:
|
|
149
|
+
return_values = [
|
|
150
|
+
# get approved evidence levels
|
|
151
|
+
[{"@rid": v} for v in APPROVED_EVIDENCE_RIDS]
|
|
152
|
+
]
|
|
153
|
+
index = -1
|
|
154
|
+
|
|
155
|
+
def __call__(self, *args, **kwargs):
|
|
156
|
+
self.index += 1
|
|
157
|
+
ret_val = self.return_values[self.index] if self.index < len(self.return_values) else []
|
|
158
|
+
return ret_val
|
|
159
|
+
|
|
160
|
+
def mock_get_source(source):
|
|
161
|
+
return {"@rid": 0}
|
|
162
|
+
|
|
163
|
+
conn = Mock(query=QueryMock(), cache={}, get_source=mock_get_source)
|
|
164
|
+
|
|
165
|
+
return conn
|
|
166
|
+
|
|
167
|
+
|
|
168
|
+
def base_graphkb_statement(disease_id: str = "disease", relevance_rid: str = "other") -> Statement:
|
|
169
|
+
statement = Statement( # type: ignore
|
|
170
|
+
{
|
|
171
|
+
"conditions": [
|
|
172
|
+
{"@class": "Disease", "@rid": disease_id, "displayName": "disease_display_name"},
|
|
173
|
+
{
|
|
174
|
+
"@class": "CategoryVariant",
|
|
175
|
+
"@rid": "variant_rid",
|
|
176
|
+
"displayName": "KRAS increased expression",
|
|
177
|
+
},
|
|
178
|
+
],
|
|
179
|
+
"evidence": [],
|
|
180
|
+
"subject": {
|
|
181
|
+
"@class": "dummy_value",
|
|
182
|
+
"@rid": "101:010",
|
|
183
|
+
"displayName": "dummy_display_name",
|
|
184
|
+
},
|
|
185
|
+
"source": None,
|
|
186
|
+
"sourceId": None,
|
|
187
|
+
"relevance": {
|
|
188
|
+
"@rid": relevance_rid,
|
|
189
|
+
"displayName": "relevance_display_name",
|
|
190
|
+
"name": "relevance_name",
|
|
191
|
+
},
|
|
192
|
+
"@rid": "statement_rid",
|
|
193
|
+
}
|
|
194
|
+
)
|
|
195
|
+
return statement
|
|
196
|
+
|
|
197
|
+
|
|
198
|
+
@pytest.fixture(autouse=True)
|
|
199
|
+
def mock_get_term_tree(monkeypatch):
|
|
200
|
+
def mock_func(*pos, **kwargs):
|
|
201
|
+
return [{"@rid": d} for d in DISEASE_RIDS]
|
|
202
|
+
|
|
203
|
+
monkeypatch.setattr(gkb_vocab, "get_term_tree", mock_func)
|
|
204
|
+
|
|
205
|
+
|
|
206
|
+
@pytest.fixture(autouse=True)
|
|
207
|
+
def mock_categorize_relevance(monkeypatch):
|
|
208
|
+
def mock_func(_, relevance_id):
|
|
209
|
+
return relevance_id
|
|
210
|
+
|
|
211
|
+
monkeypatch.setattr(gkb_statement, "categorize_relevance", mock_func)
|
|
212
|
+
|
|
213
|
+
|
|
214
|
+
class TestConvertStatementsToAlterations:
|
|
215
|
+
def test_disease_match(self, graphkb_conn, mock_get_term_tree) -> None:
|
|
216
|
+
statement = base_graphkb_statement(DISEASE_RIDS[0])
|
|
217
|
+
result = convert_statements_to_alterations(
|
|
218
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
219
|
+
)
|
|
220
|
+
|
|
221
|
+
assert len(result) == 1
|
|
222
|
+
row = result[0]
|
|
223
|
+
assert row["kbVariantId"] == "variant_rid"
|
|
224
|
+
assert row["kbStatementId"] == "statement_rid"
|
|
225
|
+
assert row["matchedCancer"]
|
|
226
|
+
assert row["kbVariant"] == "KRAS increased expression"
|
|
227
|
+
assert row["relevance"] == "relevance_display_name"
|
|
228
|
+
|
|
229
|
+
def test_no_disease_match(self, graphkb_conn) -> None:
|
|
230
|
+
statement = base_graphkb_statement("other")
|
|
231
|
+
result = convert_statements_to_alterations(
|
|
232
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
233
|
+
)
|
|
234
|
+
|
|
235
|
+
assert len(result) == 1
|
|
236
|
+
row = result[0]
|
|
237
|
+
assert not row["matchedCancer"]
|
|
238
|
+
|
|
239
|
+
def test_multiple_disease_not_match(self, graphkb_conn) -> None:
|
|
240
|
+
statement = base_graphkb_statement("disease")
|
|
241
|
+
statement["conditions"].append(
|
|
242
|
+
{"@class": "Disease", "@rid": "other", "displayName": "disease_display_name"} # type: ignore
|
|
243
|
+
)
|
|
244
|
+
result = convert_statements_to_alterations(
|
|
245
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
246
|
+
)
|
|
247
|
+
|
|
248
|
+
assert len(result) == 1
|
|
249
|
+
row = result[0]
|
|
250
|
+
assert not row["matchedCancer"]
|
|
251
|
+
|
|
252
|
+
def test_biological(self, graphkb_conn) -> None:
|
|
253
|
+
statement = base_graphkb_statement()
|
|
254
|
+
statement["relevance"]["@rid"] = "biological"
|
|
255
|
+
|
|
256
|
+
result = convert_statements_to_alterations(
|
|
257
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
258
|
+
)
|
|
259
|
+
assert len(result) == 1
|
|
260
|
+
row = result[0]
|
|
261
|
+
assert row["category"] == "biological"
|
|
262
|
+
|
|
263
|
+
def test_prognostic_no_disease_match(self, graphkb_conn) -> None:
|
|
264
|
+
statement = base_graphkb_statement()
|
|
265
|
+
statement["relevance"]["@rid"] = "prognostic"
|
|
266
|
+
|
|
267
|
+
result = convert_statements_to_alterations(
|
|
268
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
269
|
+
)
|
|
270
|
+
assert len(result) == 0
|
|
271
|
+
|
|
272
|
+
def test_prognostic_disease_match(self, graphkb_conn) -> None:
|
|
273
|
+
statement = base_graphkb_statement(DISEASE_RIDS[0])
|
|
274
|
+
statement["relevance"]["@rid"] = "prognostic"
|
|
275
|
+
|
|
276
|
+
result = convert_statements_to_alterations(
|
|
277
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
278
|
+
)
|
|
279
|
+
assert len(result) == 1
|
|
280
|
+
row = result[0]
|
|
281
|
+
assert row["category"] == "prognostic"
|
|
282
|
+
|
|
283
|
+
def test_diagnostic(self, graphkb_conn) -> None:
|
|
284
|
+
statement = base_graphkb_statement()
|
|
285
|
+
statement["relevance"]["@rid"] = "diagnostic"
|
|
286
|
+
|
|
287
|
+
result = convert_statements_to_alterations(
|
|
288
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
289
|
+
)
|
|
290
|
+
assert len(result) == 1
|
|
291
|
+
row = result[0]
|
|
292
|
+
assert row["category"] == "diagnostic"
|
|
293
|
+
|
|
294
|
+
@patch("pori_python.ipr.ipr.get_evidencelevel_mapping")
|
|
295
|
+
def test_unapproved_therapeutic(self, mock_get_evidencelevel_mapping, graphkb_conn) -> None:
|
|
296
|
+
mock_get_evidencelevel_mapping.return_value = {"other": "test"}
|
|
297
|
+
|
|
298
|
+
statement = base_graphkb_statement()
|
|
299
|
+
statement["relevance"]["@rid"] = "therapeutic"
|
|
300
|
+
statement["evidenceLevel"] = [{"@rid": "other", "displayName": "level"}] # type: ignore
|
|
301
|
+
|
|
302
|
+
result = convert_statements_to_alterations(
|
|
303
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
304
|
+
)
|
|
305
|
+
assert len(result) == 1
|
|
306
|
+
row = result[0]
|
|
307
|
+
assert row["category"] == "therapeutic"
|
|
308
|
+
|
|
309
|
+
@patch("pori_python.ipr.ipr.get_evidencelevel_mapping")
|
|
310
|
+
def test_approved_therapeutic(self, mock_get_evidencelevel_mapping, graphkb_conn) -> None:
|
|
311
|
+
mock_get_evidencelevel_mapping.return_value = {APPROVED_EVIDENCE_RIDS[0]: "test"}
|
|
312
|
+
|
|
313
|
+
statement = base_graphkb_statement()
|
|
314
|
+
statement["relevance"]["@rid"] = "therapeutic"
|
|
315
|
+
statement["evidenceLevel"] = [{"@rid": APPROVED_EVIDENCE_RIDS[0], "displayName": "level"}] # type: ignore
|
|
316
|
+
|
|
317
|
+
result = convert_statements_to_alterations(
|
|
318
|
+
graphkb_conn, [statement], "disease", {"variant_rid"}
|
|
319
|
+
)
|
|
320
|
+
assert len(result) == 1
|
|
321
|
+
row = result[0]
|
|
322
|
+
assert row["category"] == "therapeutic"
|
|
323
|
+
|
|
324
|
+
|
|
325
|
+
class TestKbmatchFilters:
|
|
326
|
+
def test_germline_kb_matches(self):
|
|
327
|
+
assert len(germline_kb_matches(GERMLINE_KB_MATCHES, GERMLINE_VARIANTS)) == len(
|
|
328
|
+
GERMLINE_KB_MATCHES
|
|
329
|
+
), "Germline variant not matched to germline KB statement."
|
|
330
|
+
assert not germline_kb_matches(
|
|
331
|
+
GERMLINE_KB_MATCHES, SOMATIC_VARIANTS
|
|
332
|
+
), "Somatic variant matched to KB germline statement."
|
|
333
|
+
assert len(germline_kb_matches(SOMATIC_KB_MATCHES, SOMATIC_VARIANTS)) == len(
|
|
334
|
+
SOMATIC_KB_MATCHES
|
|
335
|
+
), "Somatic variant not matched to somatic KB statement."
|
|
336
|
+
assert not germline_kb_matches(
|
|
337
|
+
SOMATIC_KB_MATCHES, GERMLINE_VARIANTS
|
|
338
|
+
), "Germline variant matched to KB somatic statement."
|